The Plant Cell, Vol. 6, 811-825, June 1994 O 1994 American Society of Plant Physiologists
Cell-Specific Regulation of Gene Expression in Mitochondria
during Anther Development in Sunflower
Christina J. Smart, Françoise Monéger, and Christopher J. Leaver
’
Department of Plant Sciences, University of Oxford, South Parks Road, Oxford, 0 x 1 3RB, United Kingdom
Mitochondrial gene expression was characterized during meiosis in sunflower anthers. In situ hybridizationexperiments
showed that there was a marked accumulation of four mitochondrial gene transcripts (atpA, atp9, cob, and rrn26) in
young meiotic cells. This pattern of transcript accumulation was only detected for mitochondrial genes and not for transcripts of two nuclear genes (atpB and ANT) encoding mitochondrial proteins or another nuclear gene transcript (25s
rRNA). lmmunolocalization studies showed that the pattern of accumulation of the protein product of the atpA gene,
the F1-ATP synthase a subunit, reflects that of the transcript. The expression of the nove1 mitochondrial orf522, which
is associated with the cytoplasmic male-sterile (CMS)phenotype, was also studied by in situ hybridiration. The orf522
transcripts were reduced in abundance in meiotic cells in the presence of fertility restorer genes. These results suggest
that mitochondrial gene expression is regulated in a cell-specific fashion in developing anthers and that the restorer
gene(s) may act cell specifically.
INTRODUCTION
In higher plants, mitochondrial function must be developmentally regulated to meet the changing respiratory demands of
different tissues. In some tissues, high respiratory demand
may be met by active mitochondrialbiogenesisto produce sufficient numbers of mitochondria and therefore energy in the
form of ATP (Warmke and Lee, 1978). Active biogenesis of mitochondria has been shown to occur in the root apical meristem
of Arabidopsis and other species where mitochondrial numbers increase from 65 to 200 per cell and respiration rates are
high (Yemm, 1965; Kuroiwa et al., 1992; Fujie et al., 1993). Mitochondrial biogenesis has also been reported to occur in
meiotic and tapetal cells during early anther development in
maize where numbers of mitochondria increase 20- to 40-fold
(Lee and Warmke, 1979). This developmental control of mitochondrial biogenesis may be partly regulated at the leve1 of
mitochondrial gene expression, as has been suggested by experiments where mitochondria isolated from different tissues
were found to synthesize different profiles of proteins (Newton
and Walbot, 1985; Moneger et al., 1994). Finnegan and Brown
(1990) proposed that this developmental regulation of mitochondrial gene expression occurs at the transcriptional and
post-transcriptional levels.
In many biological systems, mutations that perturb the normal developmental pathway can be used to define the
important processes occurring in that pathway. Unfortunately,
our understanding of how mitochondrial function and gene ex-
To whom correspondence should be addressed.
pression are developmentally regulated in higher plantshas
been limited because only two mitochondrial mutations have
been associated with nonlethal phenotypes: nonchromosomal
stripe (NCS) in maize and cytoplasmic male sterility (CMS) in
a range of higher plants (Newton, 1988; Hanson, 1991). Characterizationof leaf development in NCS mutants has indicated
that mitochondrial function is necessary for normal development of chloroplasts and acquisition of photosynthetic
competence (Newton and Coe, 1986; Rousell et al., 1991). Similarly, investigation of CMS has shown that mitochondrial
function is essential during pollen formation and therefore sexual reproduction in higher plants (Hanson, 1991).
CMS is a maternally inherited phenotype characterized by
an inability to produce functional pollen (Laser and Lestern,
1972). Interestingly,it only disrupts pollen production; vegetative development and female fertility are apparently unaffected.
In nature, CMS can arise spontaneously, where it may be maintained by an increase in female fertility (Couvet et al., 1990),
but for commercial purposes it is often induced by interspecific
and intraspecific crosses that introduce a nucleus into a foreign cytoplasm. In many species, the CMS phenotype is
associated with mutations in the mitochondrial genome and
can be suppressed by the action of nuclear-encoded fertility
restorer genes (Hanson, 1991). Thus, the study of fertile, CMS,
and restored fertile hybrid plants provides a good experimental system for the investigationof the developmental regulation
of mitochondrial gene expression by nuclear genes.
A great deal of our present understanding of the molecular
and biochemical basis of CMS comes from studies of male
sterility in maize (CMS-T) and petunia. In these systems,
812
The Plant Cell
polypeptide products of novel open reading frames (ORFs)
have been found to be associated with CMS (T-urf-13 and the
pcf gene, respectively) and are present at reduced levels in
the restored hybrid plants (Dewey et al., 1987; Nivison and
Hanson, 1989). However, our understandingof how these novel
polypeptides might cause pollen forming cells to abort is still
limited. It is also obvious from cytological studies of different
types of CMS within the same species that the mechanisms
of pollen abortion are diverse (Laveau et al., 1989). Therefore,
analysis of different CMS systems will extend our knowledge
of the important cytoplasmic processes occurring during pollen production.
Recently, molecular characterization of CMS in sunflower
has shown that as in the maize CMS-T and petunia systems,
a polypeptide that is the product of a novel ORF is associated
with pollen cell abortion. The abundance of this protein is reduced specifically in male florets upon restoration of fertility
(Monéger et al., 1994). CMS in sunflower provides an attractive system for further analysis because it is associated with
the expression of a single novel ORF, and the timing of meiotic
cell abortion has been characterized (Horner, 1977; Laveau
et al., 1989; Laver et al., 1991). We have undertaken a cytological investigation of anther development in sunflower florets,
coupled with further molecular analysis of CMS in sunflower,
to characterize the nature of mitochondrial gene expression
and function during anther development and the cause of
meiotic cell (meiocyte) abortion in male sterile sunflowers.
CMS in sunflower was first identified by Leclercq (1969) in
the progeny of an intraspecific cross between Helianthus annuus and H. petiolaris (PET1 type). Leclercq (1984) later
identified two dominant restorer genes of the CMS phenotype
in sunflower (Rfl and Rf2) and postulated from restoration
genetics that Rf2 is dominant in most sterile cytoplasms. In
sunflower, the CMS phenotype is associated with a mutation
in the mitochondrialgenome of the sterile line that is thought
to have been created by inversion/insertion events involving
recombination across a small repeat (Siculella and Palmer,
1988; Kohler et al., 1991; Laver et al., 1991). The insertion of
a novel fragment of DNA has led to the creation of a novel open
reading frame (orf522) downstream of the arpA gene (Kohler
et al., 1991; Laver et al., 1991). In sterile and restored hybrid
lines, orf522 is cotranscribed with atpA on a 3-kb transcript
that is translated to produce a novel 15-kD polypeptide
(ORF522), in addition to the a subunit of the mitochondrial
F1-ATPsynthase, encoded by afpA (Monéger et al., 1994). The
restorer gene(s) acts specifically in male florets to reduce
the abundance of the atpA-orf522 cotranscript and therefore
the 15-kD ORF522 polypeptide, suggesting that the expression of orf522 is probably causally related to the CMS
phenotype (Monéger et al., 1994).
In this study, we have addressed the problem of why the
expression of orf522 only affects microsporogenesis by observing the expression of mitochondrial genes during anther
development in sunflower. We showed that the atpA transcript
and the a subunit protein are particularly abundant in young
meiocytes of all lines, implying a cell-specific regulation of
expression of the mitochondrial genome. Our data indicated
that this pattern of expression is not a general phenomenon
but is only observed for mitochondriallyencoded genes. Finally, we present evidence that suggests that the restorer
gene(s) may act cell specifically in meiocytes of the restored
hybrid line to reduce the abundance of the atpA-orf522 transcript. These results provide compelling evidence for
cell-specific regulation of the expression of the mitochondrial
genome during anther development and demonstrate that nuclear genes can modify this expression.
RESULTS
Meiosis in Sunflower Anthers
To investigate the importance of mitochondrial gene expresSion during the early stages of anther development and its
significance for the CMS phenotype, RNA in situ hybridization analysis was used to investigate the spatial distribution
and abundance of mitochondrial gene transcripts. This first
required a characterization of male meiosis in sunflower
anthers.
Sunflower (H. annuus) is a member of the Compositae family, which is characterized by a single terminal inflorescence
consisting of 700 to 3000 individual disc flowers or florets
(Knowles, 1978). Sunflower florets are hermaphroditic, as they
are composed of both male and female organs. The anthers
are present in the upper part of the floret (“male floret,” represented by shaded ellipses in Figure 1C); the ovaries are present
in the lower part of the floret (“female floret,” represented by
open circles in Figure 1C). Sunflower florets exhibit protandry, that is, the male part of the flower (anthers) matures before
the female. Therefore, under normal conditions, male meiosis occurs before the sunflower inflorescence opens when the
inflorescencebud is between 2.5 and 4.5 cm in diameter (Figure
1A). Within the inflorescence, the florets develop sequentially
in whorls from the periphery to the center at a rate of about
one to four whorls a day (Hernández and Green, 1993). This
sequential maturation of disc flowers is demonstrated in the
open inflorescence shown in Figure 16.
For the purposes of this study, six stages of meiosis were
identified and investigated, beginning with the formation of the
pollen mother cells prior to meiocyte abortion in sterile anthers
and ending with microspore release from the tetrad in the fertile and restored hybrid lines (Figures 1C and 2). These phases
of meiosis were characterized by using histological and 4:6diamidino-2-phenylindole(DAPI) stains on fixed and embedded florets. It was found that the meiotic stage is correlated
with the length of the male part of the floret bud (male floret),
as shown in Figure 1C. These stages of sunflower microsporogenesis correspond to those described by Horner (1977).
The development of anthers in the fertile and sterile lines
was compared at the six stages of meiosis described above.
Development of anthers in florets of the sterile line appears
Cell-Specific Expression of Mitochondrial Genes
813
disc flowers
floret bud size
(mm)
PREMEIOSIS
LEPTOTENE
PACHYTENE
TETRAD
YOUNG
MICROSPORE
Figure 1. Sunflower Inflorescence Development.
(A) Male meiosis occurs in disc flower (floret) anthers prior to the opening of the inflorescence when the flower head is 2.5 to 4.5 cm in diameter
(at 40 days after germination).
(B) Florets develop sequentially in whorls from the periphery of the inflorescence to the center. Shown here are mature anthers in the open fertile
flower (50 days old).
(C) Diagram illustrating the six stages of meiosis characterized in this study (corresponding to those identified by Horner, 1977). During these
stages, abortion of meiocytes in anthers of the male-sterile line occurs (see Figure 2). The upper part of the floret (shaded ovals) contains the
anthers and can be described as the male floret. The lower part of the floret (open circles) represents the female floret. The length of the male
floret bud increases in a linear fashion during development and can therefore be correlated with meiotic stage.
814
The Plant Cell
FERTILE
Stage
STERILE
•y.
1. Premeiosis
2. Leptotene
3. Pachytene
4. Divisions
5. Tetrad
6. Young
Microspore
Figure 2. Anther Development in Florets of the Fertile and Sterile Sunflower Lines.
Six stages of anther development in the fertile line and the corresponding development of anthers in the sterile line showing meiocyte cell abortion
were identified by histological staining. Leptotene is the stage prior to meiocyte abortion that is used subsequently to compare all lines. Pachytene
is the first stage at which cytological abnormalities become apparent in the meiocytes of anthers of the sterile line (as observed by electron microscopy). The tapetum starts to degenerate during late pachytene in the anthers of the sterile line. M, meiocyte cells; T, tapetal cells; D, tetrads;
S, microspores. Bars = 50 jim.
Cell-Specific Expression of Mitochondrial Genes
completely normal during the premeiosis and leptotene stages
(Figure 2), but by pachytene (stage 3), abnormalities are detected in the meiocyte cells of the sterile line when observed
by electron microscopy (Laveau et al., 1989; C.J. Smart, data
not shown). By stage 4 (divisions), the tapetum shows abnormal development and has completely degenerated by the tetrad
stage in the sterile line (Figure 2). Therefore, for the purposes
of this study, comparisons of mitochondrial gene expression
in fertile, sterile, and restored hybrid lines were made between
male florets or anthers at the leptotene stage (stage 2) prior
to meiocyte abortion in the sterile line.
atpA
011873
FERTILE •
atpA
orfS22
STERILE -
Of/522 probe
etpA probe
•equwicedlvergence
B
Green
cotyledons
Etiolated
cotyledon*
Root tips
M«to
florate
' F S H " F S H " F S H ' 'F s H
Expression at the atpA Locus and the CMS Phenotype
Previously, we showed that the steady state levels of the chimeric atpA-orf522 cotranscript and the sterile specific ORF522
protein are reduced in male florets from the restored hybrid
line, indicating that the expression of O/7522 is correlated with
the CMS phenotype (Mon6ger et al., 1994). Transcripts corresponding to the orf873 (Figure 3A; previously called ORFb,
Laver et al., 1991) were not detected in RNA from different tissues by RNA gel blot analysis (data not shown), suggesting
that the disruption of orf873 is probably not causally related
to the CMS phenotype.
To further check that the reduction in the abundance of the
3-kb atpA-orf522 transcript is specific to male floret tissue, an
atpA gene-specific probe (Figure 3A) was hybridized to total
RNA extracted from green cotyledons, etiolated cotyledons,
root tips, and leptotene stage male florets of all lines in an RNA
gel blot analysis. The results are shown in Figure 3B; in the
fertile line, the atpA gene is transcribed as a 1.9-kb transcript,
whereas sterile and restored hybrid lines contain an additional
3-kb atpA-orf522 transcript (Laver et al., 1991). As shown previously, the 3-kb atpA-orf522 cotranscript is specifically
reduced in abundance in male florets of the restored hybrid
line (Figure 36; Moneger et al., 1994). In contrast, the steady
state level of the 1.9-kb atpA gene transcript remains relatively
constant between the sterile and restored hybrid lines. These
results confirm that the reduction of the 3-kb atpA-orf522
cotranscript is specific to restored hybrid male florets and does
not occur in other tissues in which there is active mitochondrial biogenesis (Figure 3B).
To investigate at which stage of anther development the
reduction of the chimeric transcript is most pronounced, an
orf522-specific gene probe (Figure 3A) was hybridized to total
RNA isolated from fertile, sterile, and restored hybrid male
florets at different stages (Figure 3C). The reduction of the
atpA-orf522 (3 kb) transcript is apparent at all stages of floret
development studied but is particularly pronounced in male
florets at the leptotene stage (stage 2) and at the tetrad/microspore stage (stages 5 and 6). The filter shown in Figure 3C
was reprobed with an atp9 gene-specific probe (Table 1) to
allow for correction of any differences in loading of RNA samples (data not shown). Densitometry was then carried out on
815
3-kb
atpA
-1.9-kb
Male florets
Root tips
1
2
3/4
5/6
Stage
F S H 'F S H"F S H"F S H"F S H'
-«-3-kb
orf522
-*-1.5-kb
Figure 3. RNA Gel Blot Analysis of atpA- and orf522-Specific Gene
Probes to Sunflower Total RNA.
(A) Diagram representing mitochondrial DNA of the fertile and sterile
lines at the atpA locus and showing the origins of the atpA- and orf522specific gene probes. The orf873 (formerly ORFb in Laver et al., 1991)
is apparently not expressed.
(B) Ten micrograms of total RNA from green cotyledons, etiolated
cotyledons, root tips, and male florets (leptotene stage 2) of fertile (F),
sterile (S), and restored hybrid (H) sunflowers was fractionated by electrophoresis, and RNA gel blot analysis was performed using an
a(p4-specific probe. The 1.9-kb (atpA) and 3-kb (atpA-orf522) transcripts
are indicated by arrows.
(C) Ten micrograms of total RNA from root tips, premeiotic stage florets
(stage 1), leptotene stage florets (stage 2), pachytene and divisions
stage florets (stages 3/4), and tetrad and young microspore stage florets
(stages 5/6) was fractionated by electrophoresis, blotted, and hybridized with the orf522-specific probe. The 3- and 1.5-kb transcripts are
indicated by arrows.
816
The Plant Cell
Table 1. Origins of Homologous and Heterologous Gene Probes Used for RNA in Situ Hybridization Experiments
Gene
Gene Product
atpA (atpl)
a
Orf522
ORF522
Subunit mitochondrial
F , Fo-ATP synthase
Species
Clone
Subclone
RNA length
(kb) in
Sunflower
H. annuus
Genomic
0.7 kb BamHllEcoRl
1.9
Laver et al.
(1991)
H. annuus
Genomic
0.25 kb BamHIlEcoRI
3.0 and 1.5
Laver et ai.
PETl
Reference
(1991)
atp9
Subunit 9 mitochondrial
F,Fo-ATP synthase
2. mays
Genomic
0.94 kb XhollBamHI
0.8
Dewey et
al. (1985)
cob
Apocytochrome b
Z. mays
Genomic
0.68 kb EcoRllHindlll
2.7
Dawson et
al. (1984)
rrn 26
Mitochondrial 26s
rRNA
T. aestivum
Genomic
0.98 kb Aval
3.3
Falconet et
al. (1985)
Z. mays
cDNA
0.35 kb BamHllSmal
2.4
Winning et
ai. (1990)
atpB (atp2)
3. ! Subunit mitochondrial
FIFo-ATP synthase
ANT
Adenine nucleotide
translocator
2. mays
cDNA
0.63 kb ClallEcoRl
2.0
Winning et
al. (1991)
Nuclear 25s rDNA
Cytoplasmic 255 rRNA
T. aestivum
Genomic
0.9 kb BamHllEcoRl
3.6
Gerlach and
Bedbrook
I19791
the autoradiograph to determine the extent of reduction of the
orf522 transcripts relative to the abundance of the atp9 transcript. There was an -75% reduction in the amount of the
chimeric 3-kb atp A-orf522 cotranscript in male florets of the
restored hybrid line at the leptotenestage of meiosis. The off522
probe also hybridized to a low abundance transcript of 1.5 kb,
which probably corresponds to the 1.2-kb orf522 transcript detected by Kohler et al. (1991). The steady state leve1 of the low
abundance, 1.5-kb orf522 transcript was also reduced in male
florets upon restoration to fertility (Figure 3C).
Cell-Specific Accumulation of Mitochondrial Gene
Transcripts in Anthers of the Fertile Line
To investigate the spatial distribution of the arpA transcripts
during normal anther development, the arpA gene probe (Figure 3A) was hybridized to anther sections of the fertile line in
situ (Figure 4). During the early stages of meiotic prophase
1 (stages 1 and 2, premeiosis and leptotene, respectively), the
arpA transcript is not distributed evenly throughout the anther
tissue, but is highly localized to the meiocyte cells. The lack
of hybridization to the leptotene stage anther sections with the
sense strand probe indicates that this is a specific signal (Figure 4). By the divisions stage (stage 4), the tapetal cells have
expanded and show high levels of accumulation of the atpA
gene transcript, and in the later tetrad and microspore stages
the arpA gene transcript is equally abundant in both cell types
and less abundant in other anther cells (Figure 4). It is thought
that the accumulation of arpA gene transcripts in the meiocyte and tapetal cells of developing anthers observed by RNA
in situ hybridization experiments accounts for the relatively high
steady state levels of atpA gene transcripts in male floret tissue observed by RNA gel blot analysis (Figure 38).
To further investigate whether the pattern of distribution of
the atp4 transcript in developing anthers is a general phenomenon for all cellular RNA or is specific to mitochondriallyencoded
genes, RNA in situ hybridizationanalysis was performed using
various gene probes. Leptotene stage (stage 2) anther sections were hybridized with probes corresponding to other
mitochondrial genes (atp9 encoding subunit 9 of the Fo-ATP
synthase, cob encoding the apocytochrome b protein, and
rm26 encoding the mitochondrial26S rRNA), nuclear genes
encoding mitochondrial proteins (ar@, encoding the p subunit
of the F1-ATP synthase, and ANT; encoding the adenine
nucleotide translocator), and a nonmitochondrial nuclear gene
(nuclear rDNA 25s encoding the 25s rRNA; see Table 1 and
Figure 5). All probes hybridized to single transcripts in total
RNA (Table 1) and hybridized specifically to anther sections
of the fertile line, as indicated by the lack of hybridization in
the sense strand probe controls (Figure 5). The intensity of
the signals with the different probes cannot be directly compared because the exposure time is different for each probe.
The results presented in Figure 5 show that the atp9, cob,
and rm26 mitochondrial gene transcripts exhibit the Same
meiocyte-specific accumulation at the leptotene stage of meiosis as the atpA gene transcript (Figure4). In contrast, the atpB
and ANT nuclear-encoded gene transcripts and the 25s rRNA
do not accumulate in meiocyte cells but appear to be more
equally abundant in all anther cell types. The mitochondrial
atp9 and rm26 gene transcripts have the same distribution
pattern as the ar@ transcript in the tapetum at the later meiotic
Cell-Specific Expression of Mitochondrial Genes
Stage
atpA
1. Premeiosis
2. Leptotene
3. Pachytene
4. Divisions
5. Tetrad
6. Young
Microspore
Sense probe
control
Figure 4. Localization of atpA Gene Transcripts in Developing Anthers
of the Fertile Sunflower Line by RNA in Situ Hybridization.
An afprt-specific, 35S-labeled antisense strand RNA probe (see Figure 3A) was synthesized and hybridized to fertile anther sections at
the six different stages of development. A sense strand probe failed
to hybridize specifically to anther sections. Pachytene is the stage at
which the meiocyte cells show the first signs of abnormal develop-
ment in the anthers of the sterile line. Bars = 50 urn.
817
stages, whereas the nuclear-encoded RNAs do not (C.J. Smart,
data not presented). These results suggest that this spatial
pattern of transcript accumulation is characteristic of genes
encoded by the mitochondrial genome and therefore is indicative of a cell-specific regulation of expression of these genes
during male meiosis.
Reduction of the atpA-orf522 Transcript Is Anther
Specific
To determine whether the novel orf522, which is associated
with CMS, exhibits the same pattern of expression as the other
mitochondrial genes, atpA, orf522, and atp9 gene transcripts
were localized in anthers of the fertile, sterile, and restored
hybrid lines by RNA in situ hybridization analysis. The atpA
and orf522gene probes (Figure 3A) were hybridized to anther
sections at the leptotene stage from the fertile, sterile, and restored hybrid lines, and an atp9 gene probe was used as a
control (Figure 6A). The atpA and atp9 mitochondrial gene transcripts accumulate in the meiocyte cells of leptotene stage
anthers of the sterile and restored hybrid lines in a similar way
to that observed in the fertile line. In leptotene stage anthers
of the sterile line, the orf522 transcripts have the same pattern of distribution as the other mitochondrial gene transcripts
studied (i.e., pronounced accumulation in the meiocyte cells).
As expected, the orf522 gene probe did not hybridize to the
anther sections of the fertile line, and the orf522 sense strand
probe did not hybridize to anther sections of the restored hybrid line, indicating that this is a specific signal.
It is evident from the data shown in Figure 6A that the extent
of hybridization of the orf522 probe to anthers of the restored
hybrid line is reduced relative to that in the anthers of the sterile line, particularly in meiocyte cells. The extent of hybridization
of the atpA probe to anther sections of the restored hybrid line
is also reduced, although the reduction is less marked than
that observed with the orf522 gene probe. In contrast, no reduction is observed for atp9 gene transcripts, which are equally
abundant in anthers of the fertile, sterile, and restored hybrid
lines. A further control is shown in Figure 6B in which the orf522
probe was hybridized to root tip sections from fertile, sterile,
and restored hybrid lines. No difference in orf522 transcript
abundance is observed between sterile and hybrid root apical meristems, confirming that the reduction in the orf522
transcript is specific to anthers of the restored hybrid lines
(Figure 6B).
Accumulation of the F,-ATP Synthase a Subunit
Protein Parallels That of the atpA Transcript
To further investigate the significance of the pattern of mitochondrial gene transcript accumulation at the protein level,
the a subunit protein of the mitochondrial F,-ATP synthase
was localized in anthers of the fertile line by fluorescent
immunolocalization.
818
The Plant Cell
atpA
trn26
cob
atp9
:ff
atp B
ANT
25 S rRNA
Figure 5. Localization of Gene Transcripts in Leptotene Stage Anthers of the Fertile Sunflower Line by RNA in Situ Hybridization.
^S-labeled antisense (A), (C), (E), (G), (I), (K), and (M) and sense (B), (D), (F), (H), (J), (L), and (N) strand probes were hybridized to leptotene
stage (stage 2) anther sections from the fertile line in situ. All probes that were used hybridized to a single transcript in RNA gel blot hybridization
experiments with total root RNA (data not presented). For the origins of gene-specific probes, see Table 1.
(A) and (B) atpA mitochondrial gene.
(C) and (D) afp9 mitochondrial gene.
(E) and (F) cob mitochondrial gene.
(G) and (H) rrn26 mitochondrial gene.
(I) and (J) atpB nuclear gene encoding a mitochondrial protein.
(K) and (L) ANT nuclear gene encoding a mitochondrial protein.
(M) and (N) 25S rRNA nuclear gene.
Bars in (A) and (B) = 50 urn.
A polyclonal antibody raised against the maize FrATP synthase a subunit protein only recognized the 58-kD a subunit
protein in a protein gel blot experiment on total male floret protein at the leptotene stage (Figure 7). This result also suggests
that male florets of the fertile, sterile, and restored hybrid lines
contain the same steady state levels of the a subunit protein.
The a subunit-specific antibody was then used to localize
the a subunit protein in anther tissue and a tetramethylrhodamine isothiocyanate-labeled fluorescent secondary antibody
was used to visualize the signal. Figures 8A to 8F show that
the a subunit protein is localized in discrete spots, probably
corresponding to individual mitochondria, in young meiocytes
and developing tapetal cells of the fertile line. In addition, it
is evident that the distribution and cell-specific accumulation
of the a subunit protein reflects that of the atpA gene transcript.
Steady state levels of the a subunit protein are apparently
highest in meiocyte cells at the leptotene stage and in tapetal
cells at the tetrad and young microspore stages. The specificity of this signal is indicated by the lack of fluorescence in the
anther sections in which the primary antibody was omitted (Figure 8H). These results suggest that the cell specificity of atpA
gene expression is regulated prior to translation.
At the divisions stage of meiosis (stage 4), the a subunit protein is localized in tight rings around the dividing meiocyte
nuclei (Figure 8D). These fluorescent rings probably correspond to the numerous mitochondria and vesicles that
Cell-Specific Expression of Mitochondrial Genes
FERTILE
Histological
Staining
STERILE
HYBRID
819
HYBRID
Sense probe control
Leptotene
Stage 2
RNA
Hybridization
atpA
*
or/522
•m •#
dtp 9
B
m
FERTILE
Histological
Staining
RNA
Hybridization
or/522
STERILE
HYBRID
HYBRID
Sense probe control
\
Figure 6. Abundance of the atpA-orf522 Transcript Is Reduced Specifically in Anthers.
(A) Localization of atpA, or/522, and atp9 gene transcripts in leptotene (stage 2) anther sections from fertile, sterile, and restored fertile hybrid
sunflowers by RNA in situ hybridization. 36S-labeled antisense strand probes hybridized to anther sections from fertile, sterile, and restored hybrid lines at the leptotene stage (stage 2). Sense strand control hybridization to anthers of the restored hybrid line shows that the antisense signal
is specific. Bar = 50 urn.
(B) Localization of orf522 gene transcripts in root tip sections from fertile, sterile, and restored fertile hybrid sunflower lines by RNA in situ hybridization.
The ^S-labeled orf522 antisense strand probe hybridized to root tips. The sense probe control is shown for root tip sections of the restored hybrid line. Bar = 100 nm.
820
The Plant Cell
male floret
protein
~F
S
58 kD
Figure 7. Protein Gel Blot Analysis of the a Subunit Protein of the
F,-ATP Synthase in Sunflower Male Floret Protein.
Protein gel blot analysis was performed on 12 ng of total protein of
male florets from fertile (F), sterile (S), and restored fertile hybrid (H)
lines using a maize a subunit-specific polyclonal antibody. The 58-kD
a subunit protein is indicated by an arrow.
surround the dividing nuclei (C.J. Smart, data not shown). Electron microscopic observations revealed that following meiotic
divisions, mitochondria become closely associated with the
nuclear envelope in tetrad and microspore cells (Figure 8G).
Immunolocalization of the a subunit protein of the FrATP
synthase was also performed on leptotene stage anther sections from fertile, sterile, and restored hybrid lines (Figures 81
to 8K). These results show that the a subunit protein accumulates in the meiocyte cells of the sterile and restored hybrid
line at the leptotene stage in a similar punctate fashion to that
observed in anthers of the fertile line. This result also confirms
that there is no apparent difference in the abundance of the
a subunit protein between fertile, sterile, and restored hybrid
lines in spite of the reduced steady state level of the 3-kb
atpA-orf522 transcript, as shown in Figure 7. These results
suggest therefore that the abundance of the a subunit protein
may be under translational or post-translational control.
DISCUSSION
In this study, we showed that mitochondria! gene expression
is regulated during development of sunflower anthers. During
the early stages of meiosis, mitochondrial gene transcripts and
proteins (the a subunit protein of the FrATP synthase) accumulate to relatively high steady state levels in the meiocyte
cells of the anther. During the later stages of meiosis, mitochondrial transcripts and proteins appear to accumulate in
tapetal cells (Figures 4 and 8). The high steady state levels
of mitochondrial gene transcripts and proteins in meiocytes
suggest that active mitochondrial biogenesis is occurring in
these cells, as has been reported in maize sporocytes and tapetal cells (Lee and Warmke, 1979). Mitochondrial biogenesis
in sunflower meiocyte cells may be the specific cellular function that is disrupted during meiosis in the male sterile line,
leading to meiocyte cell abortion. Accumulation of atpA gene
transcripts and the a subunit protein in tapetal cells of the fertile line occurring after the first signs of abnormal development
in meiocyte cells was observed in the sterile line and is therefore unlikely to be related to the CMS phenotype (Figures 4
and 8).
In previous studies, Young and Hanson (1987) found a fourto fivefold increase in the abundance of a specific pcf transcript in the anthers of the CMS line of petunia, indicating that
there is a differential developmental regulation of the level of
expression of the CMS-associated pcf mitochondrial gene. In
addition, a comparison of the translational profiles of proteins
synthesized by mitochondria isolated from different tissues of
maize (Newton and Walbot, 1985) and sunflower (Moneger et
al., 1994) suggests that certain mitochondrial genes are
differentially expressed during development. The developmental regulation of mitochondrial gene expression reported here
may be particularly significant because it is cell specific to meiocyte and tapetal cells (Figures 4 and 8).
Our results demonstrated an accumulation of both mitochondrial gene transcripts and mitochondrial proteins in meiocyte
and tapetal cells, implying that cell-specific mitochondrial gene
expression is regulated prior to translation at the level of mitochondrial gene copy number or at the transcriptional or
post-transcriptional levels (Figures 4 and 8). This increase in
transcript and protein abundance in meiocytes could be due
to a cell-specific replication of the mitochondrial genome (mitochondrial gene copy number). Alternatively, the rate of
transcription of mitochondrial genes could be up-regulated cell
specifically (transcriptional), or there may be cell-specific regulation of the stability of mitochondrial gene transcripts
(post-transcriptional).
There is also apparently some degree of translational regulation of expression of the mitochondrially encoded genes. This
is suggested by RNA gel blot analysis (Figure 36), RNA in situ
hybridization (Figure 6A), and protein gel blot analysis (Figure
7), which indicate that leptotene stage male florets of the restored hybrid line contain the same amount of the a subunit
protein as male florets of the fertile and sterile lines, despite
having only ~66% of the steady state level of atpA gene transcripts. These results also suggest that a reduced amount of
the a subunit protein of the F,-ATP synthase in male florets
of the sterile line is not responsible for meiocyte cell abortion.
If the accumulation of mitochondrially encoded gene transcripts in anther cells is associated with mitochondrial
biogenesis, then one would expect there to be a coordinated
cell-specific regulation of expression of nuclear genes encoding mitochondrial proteins. Nuclear gene transcripts of the
FrATP synthase p subunit and the adenine nucleotide
Figure 8. Immunolocalization of the Mitochondrial a Subunit Protein of the FrATP Synthase in Sunflower Anther Loculi.
Fluorescent immunolocalization of the a subunit protein was performed on anther sections using an a subunit-specific antibody. A tetramethylrhodamine isothiocyanate-labeled secondary antibody was used to reveal the position of the a subunit protein, which is indicated by the yellow
fluorescent spots that decorate the mitochondria.
(A) Premeiosis stage.
(B) Leptotene stage.
(C) Pachytene stage.
(D) Divisions stage.
(E) Tetrad stage.
(F) Microspore stage in anthers of the fertile line.
(G) Electron micrograph of a tetrad cell of the fertile line showing mitochondrial association with the nuclear envelope. M, mitochondria; N, nucleus.
(H) Control immunolocalization in leptotene stage anthers of the fertile line where the a subunit primary antibody was omitted.
(I) Leptotene stage anther of the fertile line.
(J) Leptotene stage anther of the sterile line.
(K) Leptotene stage anther of the restored fertile hybrid line.
In (G), bar = 1 urn; all other bars = 20 nm.
822
The Plant Cell
translocator (atpi3 and ANT respectively) do not show the same
cell-specific accumulation (at all meiotic stages) as observed
for the mitochondrial gene transcripts (Figure 5). Therefore,
if there was a coordinated regulation of expression of nuclear
genes encoding mitochondrial proteins, it would have to occur at the translational or post-translational levels. It has recently
been shown by tissue printing that the nuclear-encoded alternative oxidase protein accumulates in the tapetum of petunia
anthers at the tetrad stage (Conley and Hanson, 1994). The
authors also present evidence for the differential expression
of mitochondrial proteins during the later stages of meiosis,
insofar as the a subunit protein of the FI-ATP synthase accumulates in the tapetal and sporogenous cells of tetrad stage
anthers, whereas the cytochrome oxidase subunit I I protein
accumulates in the subepidermal anther tissue (Conley and
Hanson, 1994). The pattern of accumulation of the a subunit
protein at the tetrad stage of petunia anther development is
in agreement with the results presented in this study. However, the authors oniy studied two stages of anther
development, and the tissue-printing technique makes it difficult to distinguish between sporogenous (meiocyte) and tapetal
cells.
In this study, we showed by electron microscopy that mitochondria become associated with the nucleus during the later
stages of meiosis in sunflower meiocytes (Figure 8). Although
this phenomenon has been described previously at the tetrad
and young microspore stages (Simonenko, 1977; Dickinson
and Potter, 1979; Laveau et al., 1989), our data suggest that
the nuclear-mitochondrial association may begin at an earlier stage during meiotic divisions (stage 4). It is possible that
this association may be a mechanism to ensure that mitochondria are distributed equally between the fOUr microspores
following meiotic divisions.
We have previously demonstrated that the restorer gene(s)
in sunflower acts specifically in male florets resulting in a posttranscriptional reduction in the stability of the 3-kb chimeric
atpA-orf522 cotranscript and consequently a reduction in the
abundance of the ORF522 protein (Moneger et al., 1994). Here,
we show that the activity of the restorer gene(s) is specific to
anthers in the restored hybrid line (Figure 6). Localization of
the orf522 transcripts by RNA in situ hybridizationshowed that
the reduction in the abundance of the atpA-orf522 transcript
is particularly pronounced in meiocyte cells of the restored
hybrid line. However, it is possible that the cells of the middle
layer and developingtapetum, which surround the meiotic cells,
also contained reduced levels of the atpA-orf522 transcript
(Figure 6A). The ORF522 protein has also been localized in
anthers of the fertile, sterile, and restored hybrid lines (C.J.
Smart, unpublished data), and preliminary immunolocalization data suggest that the reduction of the ORF522 protein in
anthers of the restored hybrid line is meiocyte specific. In light
of these results, we suggest that the restorer gene(s) acts specifically in meiocytes during the early stages of meiosis to
reduce the stability of the chimeric atpA-orf522 transcript and
consequently the abundance of the ORF522 protein. However,
the precise specificity and mode of action of the restorer gene(s)
will only be confirmed once the gene(s) has been cloned and
its transcript and protein product are localized in situ.
Our data provide evidence that the dominant nuclear fertility restoration gene(s) aCtS anther specifically to reduce the
abundance of a chimeric transcript and polypeptide thought
to be causally related to the CMS phenotype. The only other
report of a tissue-specific action of a nuclear restorer gene
is in bean CMS where Johns et al. (1992) reported that the
action of the Fr restorer gene causes the specific loss of the
pvs region of the mitochondrial genome during pollen
development.
A considerable body of evidence now exists linking mutations in the mitochondrial genome with the failure to produce
pollen, yet why vegetative development and female fertility are
unaffected remains to be determined (Hanson, 1991). This developmental specificity may be partly due to the fact that
mutations in the mitochondrial genome, associated with CMS,
apparently do not disrupt essential mitochondrial genes (Levings,
1993). Indeed, in NCS mutants, where essential mitochondrial
genes are disrupted, the development of the whole plant is
severely affected (Newton et al., 1990; Hunt and Newton, 1991;
Rousell et al., 1991). The tissue specificity of phenotypes associatedwith mutations in human mitochondrial DNA is thought
to be conferred by a combination of tissue-specific accumulation of mutant genomes and a tissue-specific sensitivity to
reduced capacity for oxidative phosphorylation (Wallace, 1992,
1993). The tissue specificity of the CMS phenotype cannot be
attributed to heteroplasmy, because male-sterile plants are apparently homoplasmic for the CMS mutations (Hanson, 1991).
It is possible, however, that respiratory demand is higher in
anther tissue than in any other tissue and that mutant mitochondria in the anthers of CMS lines cannot support these
high demands (Singh and Brown, 1991; Levings, 1993).
Our findings provide an alternative explanation for the tissue specificity of the CMS phenotype and are consistent with
the suggestion that a high leve1of mitochondrialgene expression and biogenesis is required in the meiocyte cells of the
anther to produce sufficient mitochondria to sustain each of
the four haploid microspore cells. The provision of mitochondria required for the development of microspore cells may be
greater than that for female meiocyte cells or mitotic cells (e.g.,
in meristems) because the male meiocyte cell must divide
equally and simultaneously into four cells, each of which becomes a male gametophyte that must exist independently of
the parent anther tissue. The expression of orf522 in malesterile lines may disrupt or impair mitochondrial biogenesis
in the meiocytes leading to cell abortion.
In the most extensively studied examples of CMS, sterile
plants contain a chimeric mitochondrial ORF that is expressed
as a novel polypeptide in all tissues studied (Dewey et al., 1987;
Nivison and Hanson, 1989; Moneger et al., 1994). In these
cases, the CMS phenotype may be directly caused by the presente of the novel polypeptide or indirectly by the effect of
expression of the chimeric gene on expression of essential
mitochondrial genes (Hanson et al., 1989). In sunflower CMS,
the ORF522 protein may be directly deleterious to meiocyte
Cell-Specific Expression of Mitochondrial Genes
cells, or the cotranscription of orf522 with the afpA gene may
indirectly affect the translation of the a subunit protein of the
F1-ATP synthase. The fact that the restorer gene(s) acts anther specifically to cause reduction of the orf522 transcript
abundance of the ORF522 protein levels (Figure 6) suggests
that the ORF522 protein itself may be directly toxic to meiocytes or that it interacts with a meiocyte-specific molecule to
impair mitochondrial function (see Flavell, 1974). However, an
indirect mechanism causing CMS cannot be ruled out. The
requirement for high levels of mitochondrial gene expression
and mitochondrial biogenesis (amplification) during meiocyte
cell development may put increased demands on the
processes of transcription and translation (Figures 4 and 8).
It is possible that the cotranscription of the atpA gene with
orf522 on the 3-kb transcript may disrupt the translation and
subsequent assembly of the a subunit protein into the Fl-ATP
synthase complex without affecting the abundance of the a
subunit protein (Figure 7).
In conclusion, our results provide compelling evidence that
the expression of the mitochondrial genome is regulated in
a cell-specific manner during sunflower anther development.
Furthermore, the fact that nuclear-encoded fertility restorer
gene(s) can specifically modify the expression of a mitochondrially encoded gene (off522)in meiocyte cells suggests that
the nucleus is intimately involved in this developmental regulation. We propose that the action of the restorer gene(s) in
meiocyte cells ata stage prior to abortion of these cells in the
male-sterile lines is significant, because it suggests that expression of the CMS-associated off522 is lethal to young
meiotic cells during sunflower anther development.
METHODS
Plant Material
Sunflower seeds were provided by Rhône Poulenc Agrochimie(Lyon,
France). Fertile (RPA842B) and sterile (RPA842A) lines are isonuclear
containing the Helianthus annuus nuclear genotype; the fertile line
carries the H. annuus cytoplasm, and the sterile line contains the
H. petiolaris cytoplasm (Leclercq, 1969). The fertile restored hybrid
(RPA844H) line results from a cross between the sterile line and a
restorer line (RPA843R) carryingthe dominant nuclear restorergenes.
Seeds were germinated following surface sterilization with 5% (vlv)
sodium hypochlorite, and plants were grown in a 16-hr 28OC day and
Ehr 24OC night regime. Cotyledonsand root tips were harvested after
4 days, flower buds were harvestedafter -40 days, and flowering occurred after 50 days (see Figure 1). Male florets consist of the upper
part of the sunflower floret that contains the anthers and were dissected
from flower buds for RNA and protein extractions.
FINA Gel Blot Analysls
Total RNA was extracted from root tips, etiolated cotyledons, green
cotyledons, and male florets according to the method of Castresana
et al. (1988).Ten microgramsof RNA was fractionated by electropho-
823
resisthrough a 3-(N-morpholino)propanesulfonicacid-buffered 1.3%
(wlv) agarose gel containing 0.6 M formaldehyde and transferred to
Hybond-N membrane (Amersham International, Buckinghamshire,
U.K.) by capillary blottingovernight. Riboprobeswere prepared by in
vitro transcription according to the manufacturer'sinstructions using
32P-UTP(Amersham) and a kit from Promega. Hybridizationwas performed overnight at 6OoC in 50% (vhr) formamide, 0.75 M NaCI, 0.15
M Tris-HCI, pH 8.0, 10 mM EDTA, O.l0/0 (w/v) SDS, 1 x Denhardt's solution (see Sambrook et al., 1989),and 0.5 mg/mL denaturedsalmon
sperm DNA. Washes were performedat 6OoCin 2 x SSC (see Sambrook et al., 1989)and 0.1% (wh) SDS for 10 min, 1 x SSC and 0.1%
(w/v) SDS for 30 min, and 0.1 x SSC and 0.1% (w/v) SDS for 10 min.
The membranes were then autoradiographedusing Du Pont Cronex
film.
Protein Gel Blot Analysls
Total protein was extracted from 1 g of male floret tissue by grinding
in 3 mL of homogenizationbuffer (18% [wlv] sucrose, 10 mM MgCI2,
100 mM Tris-HCI, pH 8.0, 40 mM 0-mercaptoethanol,0.1% [w/v] SDS,
1 mM phenylmethylsulfonylfluoride, 1 mM amino caproic acid). The
homogenate was then filtered and centrifuged at 10,OOOgfor 15 min
at 4OC. To each 500 pL of supernatant, 300 pL of stop dye was added
(1% [wlv] SDS, 0.1% [w/v] bromophenolblue, 10 mM EDTA, 20Vo [w/v]
Ficoll).
To produce the a subunit polyclonal antibody, the a subunit of the
F,Fo-ATP synthase was purified from Zea mays mitochondriaaccording to the method of Hack and Leaver (1983) and electroeluted from
an SDS-polyacrylamide gel, and antibodies were raised in rabbits.
Proteins were separated by SDS-PAGE and electroblotted onto a
polyvinylidene difluoride membrane (Bio-Rad) for 1 hr at 1 A in electrophoresis buffer containing20% (v/v) methanol.The membranewas
then incubated for 15 min in TBST-BSA buffer (10 mM Tris-HCI, pH
8.0, 150 mM NaCI, 0.05% [wh] Tween 20, and 1% [wh] lyophilized BSA)
and subsequently incubated for 3 hr in a 1:10,000dilution of primary
antibody. The filter was then washed in TBST and incubated for 1 hr
with an alkaline phosphatase-conjugated secondary antibody and
washed in TBST. The antibody was detected using an alkaline phosphatase (AP) color reaction, whereby the filter was equilibrated in 10
mL of AP buffer (100 mM Tris, pH 9.5, 100 mM NaCI, 5 mM MgC12)
and then incubated in AP color solution (10 mL AP buffer, 66 pL nitro
blue tetrazolium [Sigma]), (0.5 g in 10 mL 70% [v/v] dimethylformamide, 33 pL 5-bromo-4-chloro-3-indolyl phosphate [Sigma]), and (0.5 g
in 10 mL 100% [vlv] dimethylformamide).The reaction was stopped
after the signal had appeared by washing the filter in water.
Preparation of Material for Light Microscopy
Plant tissues were dissected,fixed, embedded, and sectionedaccording to the method described by Langdale (1993).
Staging of Male Meiosis
The stage of meiosis was determined by the degree of condensation
of chromatin after staining with 4',6-diamidino-2-phenylindole(DAPI)
(Sigma). Paraplast wax was removed from sections by incubation in
cnp 30 (Penetone, Cramlington, U.K.) for 2 x 10 min, and sections
were rehydratedthrough an EtOH series. Sections were then incubated
in O.l0/0 (whr) BSA in PBS (10 mM sodium phosphate, pH 7.5, 150 mM
824
The Plant Cell
NaCI) for 10 min followed by 20 min in 200 pL of 0.lYo (wh) DAPl in
0.01% (vh) Tween 20-PBS. The sections were then washed in BSA-PBS
and mountedin Citifluor (Agar Scientific,Stansted, U.K.). Sections were
viewed using a fluorescentmicroscope(Axiophot; Zeiss, Oberkochen,
Germany) at an excitation wavelength of 365 nm.
Histological staining of sections was performed according to the
method described by Langdale (1993).
Liddell and Cledwyn Merriman for their technical assistance and to
John Baker for help with the figure preparation. This work was supported by the Bio Avenir programme, supported by Rhane-Poulenc
with the participation of the French Ministryof Researchand the French
Ministryof Industry. F.M. was in receipt of European Economic Community and Glasstone Fellowships.
Received February 14, 1994; accepted April 26, 1994.
RNA In Sltu Hybridlzation
RNA in situ hybridization was performed using 35S-UTP (Du
Pont-New England Nuclear)-labeled RNA probes according to the
method of Langdale (1993). All clones used were subcloned into
pBluescript,pBS KS+, or pGEM vectors so that sense and antisense
RNA probescould be synthesized by in vitro transcription. The origin
and size of these probes are shown in Table 1. For each experiment,
sections from florets at different stages or different lines were present
on the same slide so that all sections were incubated with the same
RNA probe and were exposed for the same lengthof time. Slides were
viewed and photographed with a Dialux (20 EB; Leitz, Wetzlar, Germany) microscope on llford PAN-F film.
lmmunolocalizatlon
Tissue sections were dewaxed and dehydrated as described previously and incubated in 0.1% (w/v) BSA-PBS for 5 min at room
temperature. Sections were then blocked using 0.1% (wlv) goat IgG
(Sigma) for 15 min at room temperature and washed for 15 min in
BSA-PBS. Sections were then incubated overnight with 100 pL of a
1:20dilution of the a subunit antibody in BSA-PBS and then washed
sequentially for 15 min in 0.5% (w/v) BSA-PBS, 0.01% (v/v) Tween 20PBS, and 0.1% (w/v) BSA-PBS. Sections were then incubated for
1 hr at room temperature with tetramethylrhodamineisothiocyanateconjugatedgoat anti-rabbit antisera (diluted in PBS as recommended
by the manufacturer; Sigma). Slides were then washed in BSA-PBS
and mounted in Citifluor. Control slides were treated identically except that sections were incubated overnight in BSA-PBS instead of
primary antibody. All slides were viewed using a fluorescent microscope (Axiophot; Zeiss) at an excitation wavelength of 550 nm and
photographed on transparency film (400; Fuji, Tokyo, Japan).
Preparation of Material for Electron Mlcroscopy
Floretswere dissected and fixed on ice according to Karnovsky (1965)
and postfixed in 1% (vh) osmium tetroxide (TAAB, Berkshire, U.K.).
Samples were dehydrated through acetone and embedded in resin
(TAAB). Sections (0.09 pm) were cut on an ultramicrotome(MT 5000;
Sorvall, Stevenage, U.K.), stainedwith uranyl acetate followed by lead
citrate (Reynolds, 1963), and obsewed with a transmisson electron
microscope (JEM 2000 EX; Jeol, Ltd., Welwyn Garden City, U.K.).
ACKNOWLEDGMENTS
We thank Jane Langdalefor her constant advice and support and Hugh
Dickinsonand Georges Freyssinet for interestingdiscussions.Thanks
also to Davey Jones for useful suggestions.We are grateful to Andrew
REFERENCES
Castresana, C,, Garcia-Luque, I., Alonso, E., Malik, V.S., and
Cashmore, A.R. (1988).60th positive and negative regulatory ele-
ments mediate expression of a photoregulated CAB gene from
Nicotiana plumbaginifolia. EMBO J. 7, 1929-1936.
Conley, C,A., and Hanson, M.R. (1994). Tissue-specific protein expression in plant mitochondria. Plant Cell 6, 85-91.
Couvet, D., Atlan, A., Belhassen, E., Gltddon, C., Gouyon, P.H.,
and Kjellberg, F. (1990).Co-evolutionbetween two symbionts:The
case of cytoplasmic male-sterility in higher plants. In Oxford Surveys in Evolutionary Biology, Vol. 7, D. Futuyma and J. Antonovics,
eds (Oxford: Oxford University Press), pp. 225-250.
Dawson, A.J., Jones, V.P., and Leaver, C.J. (1984). The
apocytochromeb gene in maize mitochondria does not contain introns and is preceded by a potential ribosome binding site. EMBO
J. 3, 2107-2113.
Dewey, R.E., Schuster, A.M., Levings, C.S., 111, and Timothy, D.H.
(1985). Nucleotide sequence of the Fo-ATPaseproteolipid(subunit
9) gene of maize mitochondria. Proc. Natl. Acad. Sci. USA 82,
1O15-101 9.
Dewey, R.E., Timothy D.H., and Levings, C.S., 111 (1987).A mitochondrial protein associated with cytoplasmic male sterility in the T
cytoplasm of maize. Proc. Natl. Acad. Sci. USA 84, 534-5376.
Dlckinson,H.G., and Potter, U. (1979).Post-meioticnucleocytoplasmic
interaction in Cosmosbipinnatus: Early events at the nuclear envelope. Planta 145, 449-457.
Falconet, D., Delorme, S., Lejeune, B., SClgnac, M., Delcher, E.,
Bazetoux, S., and QuBtier, F. (1985). Wheat mitochondrial 26s
ribosomal RNA gene has no intron and is present in multiple copies arising by recombination. Curr. Genet. 9, 169-174.
Finnegan, P.M., and Brown, G.G. (1990). Transcriptional and posttranscriptionalregulation of RNA levels in maize mitochondria. Plant
Cell 2, 71-83.
Flavell, R. (1974). A model for the mechanism of cytoplasmic male
sterility in plants with special reference to maize. Plant Sci. Lett.
3, 259-263.
Fujle, M., Kuroiwa, H., Kawano, S.,and Kuroiwa, T. (1993).Studies
on the behavior of organelles and their nucleoids in the root apical
meristem of Ambidopsis tbaliana (L.) Col. Planta 189, 443-452.
Gerlach, W.L., and Bedbrook, J.R. (1979). Cloning and characterization of ribosomal RNAgenesfrom wheat and barley. Nucl.Acids
Fies. 7, 1869-1885.
Hack, E., and Leaver, C.J. (1983). The a-subunit of the maize FqATPase is synthesizedin the mitochondrion.EMBOJ. 2,1783-1789.
Hanson, M.R. (1991). Plant mitochondrial mutations and male sterility. Annu. Rev. Genet. 25, 461-486.
Cell-Specific Expression of Mitochondrial Genes
Hanson, M.R., Pruitt, K.D., and Nivison, H.T. (1989). Male sterility
loci in plant mitochondrial genomes. In Oxford Surveys of Plant Molecular and Cell Biology, Vol. 6, B.J. Miflin, ed (Oxford: Clarendon
Press), pp. 61-85.
Hernhdez, L.F., and Green, P.B. (1993). Transductions for the expression of structural pattern: Analysis in sunflower. Plant Cell 5,
1725-1738.
Horner, H.T. (1977). A comparativelight-and electron-microscopicstudy
of microsporogenesis in male-fertile and cytoplasmic male-sterile
sunflower (Helianthus annuus). Am. J. Bot. 64, 745-759.
Hunt, M.D., and Newton, K.J. (1991). The NCS3 mutation: Genetic
evidence for the expressionof ribosomal protein genes in Zea mays
mitochondria. EMBO J. 10, 1045-1052.
Johns, C., Lu, M., Lyznik, A., and Mackenzie, S.(1992). A mitochondrial DNA sequence is associatedwith abnormal pollen development
in cytoplasmic male sterile bean plants. Plant Cell 4, 435-449.
Karnovsky, M.J. (1965). Aformaldehyde-glutaraldehydefixative of high
osmolarityfor use in electron microscopy.J. Cell Biol. 27,137A-138A.
Knowles, P.F. (1978). Morphologyand anatomy. In Sunflower Science
and Technology, Agronomy Ser. No. 19, J.F. Carter, ed (Madison,
WI: American Society of Agronomy), pp. 55-87.
Kohler, R.H., Horn, R., Wssl, A., and Zetsche, K. (1991). Cytoplasmic male sterility in sunflower is correlated with the co-transcription
of a new open readingframe with the atpA gene. MOI.Gen. Genet.
227, 369-376.
Kurolwa, T., Fujie, M., and Kuroiwa, H. (1992). Studies on the behavior of mitochondrialDNA: Synthesisof mitochondrialDNAoccurs
actively in a specific region just above the quiescent center in the
root meristem of Fielargoniumzonale. J. Cell Sci. 101, 483-493.
Langdale, J.A. (1993). In situ hybridization.In The Maize Handbook,
M. Freeling and V. Walbot, eds (New York: Springer-Verlag), pp.
165-180.
Laser, K.D., and Lestern, N.R. (1972). Anatomy and cytology of
microsporogenesis in angiosperms. Bot. Rev. 38, 425-454.
Laveau, J.H., Schneider, C., and Berville, A. (1989). Microsporogenesis abortion in cytoplasmic male sterile plants from H. pefiolaris
or H. petiolarisfallax crossed by sunflower (Helianthusannuus).Ann.
BOt. 64, 137-148.
Laver, H.K., Reynolds, S.J., MonBger, F., and Leaver, C.J. (1991).
Mitochondrial genome organizationand expressionassociatedwith
cytoplasmic male sterility in sunflower (Helianthus annuus). Plant
J. 1, 185-193.
Leclercq, P. (1969). Une st6rilit6 male cytoplasmic chez le tournesol.
Ann. AméI. Plantes 19, 99-106.
Leclemq, P. (1984). ldentification de gbnes de restauration de fertilité
sur cytoplasmes st6rilisants chez le tournesol. Agronomie 4,573-576.
Lee, S.-L.J., and Warmke, H.E. (1979). Organelle size and number
in fertile and Tqtoplasmic malesterile corn. Am. J. Bot. 66,141-148.
Levings, CS., 111 (1993). Thoughts on cytoplasmic male sterility in cms-7
maize. Plant Cell 5, 1285-1290.
MonBger, F., Smart, C.J., and Leaver, C.J. (1994). Nuclear restoration of cytoplasmic male sterility in sunflower is associated with the
tissue-specific regulation of a novel mitochondrialgene. EMBO J.
13, 8-17.
825
Newton, K.J. (1988). Plant mitochondrialgenomes: Organization, expression and variation. Annu. Rev. Plant Physiol. MOI. Biol. 39,
503-532.
Newton, K.J., and Coe, E.H. (1986). MitochondrialDNA changes in
abnormal growth (non-chromosomalstripe) mutantsof maize. Proc.
Natl. Acad. Sci. USA 83, 7363-7366.
Newton, K.J., and Walbot, V. (1985). Maize mitochondriasynthesize
organ-specific polypeptides. Proc. Natl. Acad. Sci. USA 82,
6879-6883.
Newton, K.J., Knudsen, C., Gabay-Laughnan, S.,and Laughnan,
J.R. (1990). An abnormal growth mutant in maize has a defective
mitochondrial cytochrome oxidase gene. Plant Cell 2, 107-113.
Nivison, H.T., and Hanson, M.R. (1989). ldentificationof a mitochondrial protein associated with cytoplasmic male sterility in petunia.
Plant Cell 1, 1121-1130.
Reynolds, E.S. (1963). The use of lead citrate at high pH as an electron opaque Stain in electron microscopy.J. Cell Biol. 17, 208-212.
Rousell, D.L., Thompson, D.L., Pallardy, S.G., Mlles, D., and
Newton, K.J. (1991). Chloroplast structure and function is altered
in the NCS2 maize mitochondrialmutant. Plant Physiol.96,232-238.
Sambmok, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual. (Cold Spring Harbor, NY: Cold Spring
Harbor Laboratory Press).
Siculella, L., and Palmer, J.D. (1988). Physical and gene organization of mitochondrial DNA in fertile and male sterile sunflower.
CMS-associatedalterationsin structureand transcriptionof the atpA
gene. Nucl. Acids Res. 16, 3787-3799.
Slmonenko, V.K. (1977). Ultrastructureof developing microspore in
sunflower. Tsitologiya I Genetika 11, 395-398.
Singh, M., and Brown, G.G. (1991). Suppression of cytoplasmic male
sterility by nuclear genes alters expressionof a novel mitochondrial
gene region. Plant Cell 3, 1349-1362.
Wallace, D.C. (1992). Mitochondrialgenetics: A paradigm for aging
and degenerative diseases? Science 256, 628-632.
Wallace, D.C. (1993). Mitochondrialdiseases: Genotype versus phenotype. Trends Genet. 9, 128-133.
Warmke, H.E., and Lee, S.-L.J. (1978). Pollen abortion in T cytoplasmic male-sterilecorn (Zeamays):A suggestedmechanism.Science
200, 561-563.
Winning, B.M., Bathgate, B., Purdue, P.E., and Leaver, C.J. (1990).
Nucleotide sequence of a cDNA encoding the beta subunit of the
mitochondrialATP synthase from Zea mays. Nucl. Acids Res. 18,
5885.
Wlnning, B.M., Day, C.D., Sarah, CJ., and Leaver, C.J. (1991). Nuclectide sequence of two cDNAs encoding the adenine nucleotide
translocator from Zea mays L. Plant MOI.Biol. 17, 305-307.
Yemm, E.W. (1965). The respirationof plants and their organs. In Plant
Physiology, A Treatise, Vol. IV-A, F.C. Steward, ed (New York: Academic Press, Inc.), pp. 231-310.
Young, E.G., and Hanson, M.R. (1987). A fused mitochondrialgene
associated with cytoplasmic male sterility is developmentally regulated. Cell 50, 41-49.
Cell-specific regulation of gene expression in mitochondria during anther development in
sunflower.
C J Smart, F Monéger and C J Leaver
Plant Cell 1994;6;811-825
DOI 10.1105/tpc.6.6.811
This information is current as of June 14, 2017
Permissions
https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X
eTOCs
Sign up for eTOCs at:
http://www.plantcell.org/cgi/alerts/ctmain
CiteTrack Alerts
Sign up for CiteTrack Alerts at:
http://www.plantcell.org/cgi/alerts/ctmain
Subscription Information
Subscription Information for The Plant Cell and Plant Physiology is available at:
http://www.aspb.org/publications/subscriptions.cfm
© American Society of Plant Biologists
ADVANCING THE SCIENCE OF PLANT BIOLOGY